Contents lists available at ScienceDirect Brain and Language journal homepage: www.elsevier.com/locate/b&l Review EEG mu rhythms: Rich sources of sensorimotor information in speech processing Tim Saltuklaroglu a, ⁎ , Andrew Bowers b , Ashley W. Harkrider a , Devin Casenhiser a , Kevin J. Reilly a , David E. Jenson c , David Thornton d a Department of Audiology and Speech-Language Pathology, University of Tennessee Health Sciences, Knoxville, TN 37996, USA b University of Arkansas, Epley Center for Health Professions, 606 N. Razorback Road, Fayetteville, AR 72701, USA c Department of Speech and Hearing Sciences, Elson S. Floyd College of Medicine, Spokane, WA 99210-1495, USA d Department of Hearing, Speech, and Language Sciences, Gallaudet University, 800 Florida Avenue NE, Washington, DC 20002, USA 1. Introduction The purpose of this review is to advance the idea that the electro- encephalographic (EEG) mu (μ) rhythm can be used to make time- sensitive measures of sensorimotor integration during speech percep- tion and production. Herein, we discuss the history of the mu rhythm and its early ties to mirror neuron research. Subsequently, we provide evidence that measurements of mu rhythms should encompass both alpha and beta frequency ranges (Buzsaki, 2006; Kane et al., 2017) since they reflect separate but dependent neurological functions, and we demonstrate that mu rhythms are sensitive to multi-faceted changes in speech-related sensorimotor processing. Although we recognize that an understanding of mu rhythm function as it relates to motor and cognitive domains is still emerging, evidence demonstrating its re- sponsiveness to movement tasks suggests its utility in studying normal and disordered speech motor control. Moreover, because the mu rhythm has also been found to be responsive during auditory perception and cognitive tasks, and because models of speech perception suggest an inherent interface between perception and action, mu rhythms may afford a unique opportunity to examine the relationships among speech perception, speech production, and cognition. Additionally, we will demonstrate that independent component analysis (ICA) identifies ro- bust mu rhythms with both alpha and beta spectral peaks. Finally, we will discuss how data obtained through time-frequency analysis of speech perception and production tasks inform speech-related sensor- imotor integration and current models of speech perception and pro- duction. 1.1. Early mu rhythm studies The EEG mu rhythm has been of interest to neuroscientists since the 1950s. Gastaut and Bert (1954) were among the first to describe its incidence, source, spectral nature and response patterns as participants watched various films. In 17 of 80 participants, they observed a rhythm with a dominant frequency in the alpha range (∼9 ± 2 Hz) that emanated from Rolandic regions as a series of asymmetric arches in raw EEG traces. The authors described the rhythm as being ‘blocked’ when participants in their experiment moved, identified with a person being shown in a film, or watched two boxers hit each other. Early studies found mu rhythms infrequently (Klass & Bickford, 1957; Koshino & Isaki, 1986; Koshino & Niedermeyer, 1975), and considered them to be evidence of neuropathology (Gastaut & Bert, 1954). However, im- provements in EEG recording and analysis techniques demonstrated that the mu rhythm can be found in infancy (Stroganova, Orekhova, & Posikera, 1999), and is observed ubiquitously in healthy adults (Kuhlman, 1978; Pfurtscheller, 1986; Tiihonen, Kajola, & Hari, 1989). 1.2. Mu rhythms and mirror neurons While early studies provided the groundwork for understanding mu response patterns, it was the subsequent discovery of mirror neurons (i.e., neurons that fire in response to both observation and action) that supplied the missing piece enabling a functional interpretation of mu rhythms. The contexts under which mirror neurons were found to be active suggest that they represent a neural substrate linking action and perception. This finding has implications for understanding a myriad of human abilities including others’ intents and actions (Oberman, Pineda, & Ramachandran, 2007), imitation (Bernier, Dawson, Webb, & Murias, 2007), empathy (Yang, Decety, Lee, Chen, & Cheng, 2009), the evolu- tion of language (Rizzolatti & Arbib, 1998), language development (Theoret & Pascual-Leone, 2002), and speech perception (Liberman & Whalen, 2000). Mirror regions in humans were identified in sensor- imotor regions including dorsal and ventral premotor cortex, somato- sensory cortex, and inferior parietal lobe (Gazzola & Keysers, 2009; Keysers, 2009; Keysers & Gazzola, 2009). They have been identified in both adults and infants (Southgate, Johnson, Osborne, & Csibra, 2009; Southgate, Johnson, El Karoui, & Csibra, 2010; Marshall & Meltzoff, 2011). Since these regions largely coincide with sources of mu rhythms, https://doi.org/10.1016/j.bandl.2018.09.005 Received 11 April 2017; Received in revised form 27 September 2017; Accepted 23 September 2018 ⁎ Corresponding author. E-mail address: tsaltukl@uthsc.edu (T. Saltuklaroglu). Brain and Language xxx (xxxx) xxx–xxx 0093-934X/ © 2018 Elsevier Inc. All rights reserved. Please cite this article as: Saltuklaroglu, T., Brain and Language, https://doi.org/10.1016/j.bandl.2018.09.005